Abstract

Introduction: The relationship between milk intake and risk of diabetes is controversial with substantial ethnical difference. Whether such difference can be partially explained by LCT genotype that varies by ethnical background is unknown. Hypothesis: Higher milk intake is associated with lower risk of diabetes only in lactase non-persistent (LNP) individuals ( LCT rs4988235, GG) but not in LP individuals (AA/AG). Methods: We examined associations of milk intake with incident diabetes (n=7089, 768 incident cases over 6 years), 490 gut microbial species (n=1767) and 624 serum metabolites (n=3110), stratified by LCT genotype in US Hispanics from the HCHS/SOL. We related the identified microbial species and metabolites with metabolic traits and/or incident diabetes. Results: Higher milk intake was associated with lower risk of diabetes in LNP but not in LP individuals (Fig A). Milk intake was associated with different microbial species by LCT genotype, including 13 species specific to LNP individuals (Fig B), especially Bifidobacterium sp. (enriched), which were favorably associated with several metabolic traits (Fig C). In LNP individuals, milk intake was specifically associated with 20 metabolites (Fig D). Metabolites positively associated with milk, especially indolepropinate and β-cryptoxanthin, were favorably associated with metabolic traits and risk of diabetes (Fig E). Opposite patterns were found for metabolites negatively associated with milk, especially for several bile acids and branched chain amino acid metabolites (Fig E). Some of these metabolites are known microbiota related, were correlated with LNP specific milk-related bacteria, and partially mediated the milk-diabetes association in LNP individuals (Fig F). Such associations were not observed for LP specific milk-related bacteria or metabolites. Conclusion LCT genotype may modify the relationship between milk intake and diabetes, with a beneficial association in LNP individuals, partially explained by gut microbiota and serum metabolites.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call